Pancreatic resection of isolated metastases from nonpancreatic primary cancers

Background: The goal of this study is to report the safety and efficacy of pancreatic resection for isolated metastatic cancers from nonpancreatic primary disease. Methods: We retrospectively identified patients from a single institution's prospectively gathered pancreaticobiliary database from 1970 to 2007 who underwent a pancreatic resection for metastatic disease. Results: Forty-nine patients were identified with metastatic lesions to the pancreas. Pancreaticoduodenectomy, distal pancreatectomy, and total pancreatectomy were performed in 31, 14, and 4 patients, respectively. Pathology distribution was as follows: 21 renal cell carcinoma (RCC), 6 gallbladder cancer, 4 lung cancer, 4 ovarian cancer, 4 sarcoma, 3 melanoma, 2 colon cancer, 1 breast cancer, 1 hepatocellular carcinoma, 1 seminoma, 1 Langerhans cell histiocytosis, and 1 nonpancreatic endocrine cancer. Postoperative morbidity was 48%. There were no perioperative deaths. A statistically significant difference in survival was found between cancer types (P = .007) with median survivals ranging from 4.8 years for RCC to .9 years for melanoma. Univariate analysis demonstrated a survival disadvantage for patients with perineural (hazard ratio [HR] = 5.4, P = .004) and vascular invasion (HR = 4.4, P = .002). The most commonly resected metastatic lesion of the pancreas was RCC. Eighteen of the 23 patients with RCC had a metachronous lesion with a median length between initial operation and pancreatic resection of 9.3 years. Metachronous lesions had a survival similar to that of synchronous lesions (HR = 1.0, P = .98). Vascular invasion (HR = 2.4, P = .007) and lymph node metastases (HR = 24.1, P = .01) were associated with greater mortality. Conclusion: Long-term survival can be achieved in patients undergoing resection of isolated metastases to the pancreas.